-
Enhancing Fingerprint Image Synthesis with GANs, Diffusion Models, and Style Transfer Techniques
Authors:
W. Tang,
D. Figueroa,
D. Liu,
K. Johnsson,
A. Sopasakis
Abstract:
We present novel approaches involving generative adversarial networks and diffusion models in order to synthesize high quality, live and spoof fingerprint images while preserving features such as uniqueness and diversity. We generate live fingerprints from noise with a variety of methods, and we use image translation techniques to translate live fingerprint images to spoof. To generate different t…
▽ More
We present novel approaches involving generative adversarial networks and diffusion models in order to synthesize high quality, live and spoof fingerprint images while preserving features such as uniqueness and diversity. We generate live fingerprints from noise with a variety of methods, and we use image translation techniques to translate live fingerprint images to spoof. To generate different types of spoof images based on limited training data we incorporate style transfer techniques through a cycle autoencoder equipped with a Wasserstein metric along with Gradient Penalty (CycleWGAN-GP) in order to avoid mode collapse and instability. We find that when the spoof training data includes distinct spoof characteristics, it leads to improved live-to-spoof translation. We assess the diversity and realism of the generated live fingerprint images mainly through the Fréchet Inception Distance (FID) and the False Acceptance Rate (FAR). Our best diffusion model achieved a FID of 15.78. The comparable WGAN-GP model achieved slightly higher FID while performing better in the uniqueness assessment due to a slightly lower FAR when matched against the training data, indicating better creativity. Moreover, we give example images showing that a DDPM model clearly can generate realistic fingerprint images.
△ Less
Submitted 20 March, 2024;
originally announced March 2024.
-
Demystifying Competition and Cooperation Dynamics of the Aerial mmWave Access Market
Authors:
Olga Galinina,
Leonardo Militano,
Sergey Andreev,
Alexander Pyattaev,
Kerstin Johnsson,
Antonino Orsino,
Giuseppe Araniti,
Antonio Iera,
Mischa Dohler,
Yevgeni Koucheryavy
Abstract:
Cellular has always relied on static deployments for providing wireless access. However, even the emerging fifth-generation (5G) networks may face difficulty in supporting the increased traffic demand with rigid, fixed infrastructure without substantial over-provisioning. This is particularly true for spontaneous large-scale events that require service providers to augment capacity of their networ…
▽ More
Cellular has always relied on static deployments for providing wireless access. However, even the emerging fifth-generation (5G) networks may face difficulty in supporting the increased traffic demand with rigid, fixed infrastructure without substantial over-provisioning. This is particularly true for spontaneous large-scale events that require service providers to augment capacity of their networks quickly. Today, the use of aerial devices equipped with high-rate radio access capabilities has the potential to offer the much needed "on-demand" capacity boost. Conversely, it also threatens to rattle the long-standing business strategies of wireless operators, especially as the "gold rush" for cheaper millimeter wave (mmWave) spectrum lowers the market entry barriers. However, the intricate structure of this new market presently remains a mystery. This paper sheds light on competition and cooperation behavior of dissimilar aerial mmWave access suppliers, concurrently employing licensed and license-exempt frequency bands, by modeling it as a vertically differentiated market where customers have varying preferences in price and quality. To understand viable service provider strategies, we begin with constructing the Nash equilibrium for the initial market competition by employing the Bertrand and Cournot games. We then conduct a unique assessment of short-term market dynamics, where two licensed-band service providers may cooperate to improve their competition positions against the unlicensed-band counterpart intruding the market. Our unprecedented analysis studies the effects of various market interactions, price-driven demand evolution, and dynamic profit balance in this novel type of ecosystem.
△ Less
Submitted 17 August, 2016;
originally announced August 2016.
-
Latent modeling of flow cytometry cell populations
Authors:
Jonas Wallin,
Kerstin Johnsson,
Magnus Fontes
Abstract:
Flow cytometry is a widespread single-cell measurement technology with a multitude of clinical and research applications. Interpretation of flow cytometry data is hard; the instrumentation is delicate and can not render absolute measurements, hence samples can only be interpreted in relation to each other while at the same time comparisons are confounded by inter-sample variation. Despite this, cu…
▽ More
Flow cytometry is a widespread single-cell measurement technology with a multitude of clinical and research applications. Interpretation of flow cytometry data is hard; the instrumentation is delicate and can not render absolute measurements, hence samples can only be interpreted in relation to each other while at the same time comparisons are confounded by inter-sample variation. Despite this, current automated flow cytometry data analysis methods either treat samples individually or ignore the variation by for example pooling the data. In this article we introduce a Bayesian hierarchical model for studying latent relations between cell populations in flow cytometry samples, thereby systematizing inter-sample variation. The model is applied to a data set containing replicated flow cytometry measurements of samples from healthy individuals, with informative priors capturing expert knowledge. It is shown that the technical variation in the inferred cell population sizes is small in comparison to the intrinsic biological variation. The large size of flow cytometry data, where a single sample can contain measurements on hundreds of thousands of cells, necessitates computationally efficient methods. To address this, we have implemented a parallel Markov Chain Monte Carlo scheme for sampling the posterior distribution.
△ Less
Submitted 13 February, 2015;
originally announced February 2015.
